˗ˏˋ DC ˎˊ˗

coder ||
games | bots | nlp | gen art
4 startups / 3 continents / 2 exits

[˗ˏˋ DC ˎˊ˗ @dcsan]

#AITinkerers NYC hackathon about GCP
debate podcaster hotakes with support from AI Joe or Lex fine tuned models.
Amazing how much difference a little fine-tune makes!
https://t.co/G5DITTLVzB

[˗ˏˋ DC ˎˊ˗ @dcsan]

@rgeade @clairevo @chatprd but does that lead to any discovery? the chatGPT app store seems very weak for that.
maybe you can direct ppl there for a free trial but then you're just sending your leads to openAI

[˗ˏˋ DC ˎˊ˗ @dcsan]

@clairevo @chatprd maybe if you provide a free trial as a chatGPT/app then people can use their own LLM account to do some of the reasoning, and just use your service for the organization part?

[˗ˏˋ DC ˎˊ˗ @dcsan]

@RobinHill85 @pokerheadrush takes one to know one poker larper!

[˗ˏˋ DC ˎˊ˗ @dcsan]

@GumVue That list is wild, line art looks crisp at native 4K

[˗ˏˋ DC ˎˊ˗ @dcsan]

@buildwithvenkey Stale state bugs are silent UI gremlins. Good call tying the editor to the active item

[˗ˏˋ DC ˎˊ˗ @dcsan]

@MoSuvo While the numbers feel modest, tracking daily keeps the grind honest

[˗ˏˋ DC ˎˊ˗ @dcsan]

@buildwithvenkey Showing local prices cuts the mental math, so the $20 feels like a concrete number instead of an abstract guess

[˗ˏˋ DC ˎˊ˗ @dcsan]

@Guptha_Gudivada Clear breakdown, especially the startup probe timing. Makes setting thresholds feel like less guesswork

[˗ˏˋ DC ˎˊ˗ @dcsan]

@zazzy_stardust While most wrappers still force you to juggle keys, this single endpoint cuts the friction down to the basics

[˗ˏˋ DC ˎˊ˗ @dcsan]

@paolo_scales The breakdown feels spot on, most B2B creators I know see the sweet spot around that 60-90 minute window

[˗ˏˋ DC ˎˊ˗ @dcsan]

@aaliya_va Sharing a walkthrough feels more like a live interview than a resume, and people actually watch the results

[˗ˏˋ DC ˎˊ˗ @dcsan]

@MuralistRNow Short tags fade, a personal story sticks like a well timed bluff

[˗ˏˋ DC ˎˊ˗ @dcsan]

@XCapitalBuilder While MU looks cheap, its growth ceiling feels limited next to the AI runway behind NVDA

[˗ˏˋ DC ˎˊ˗ @dcsan]

@AamirAnsar94694 Good rundown, helps me see where a harness ends and a skill begins

[˗ˏˋ DC ˎˊ˗ @dcsan]

@Mitchell_Suen Most of the launch grind ends up in tweaking auth edge cases and monitoring latency spikes

[˗ˏˋ DC ˎˊ˗ @dcsan]

@emaann28 @testdriveso The leaderboard is a nice badge, but the real payoff is the stash of tools you’ve added to your toolbox

[˗ˏˋ DC ˎˊ˗ @dcsan]

@Alora_Tech_Ai0 While the alternatives are straightforward, getting legacy systems to play nice is still a marathon, not sprint

[˗ˏˋ DC ˎˊ˗ @dcsan]

@_PradeepGoel @GoogleDeepMind Turning a years long effort into four weeks makes formal verification feel more within reach for ordinary teams

[˗ˏˋ DC ˎˊ˗ @dcsan]

@MaximumADHD While the hype drags everyone into trashy demos, the real win is shaving hours off build scripts and letting us think

[˗ˏˋ DC ˎˊ˗ @dcsan]

@wespreadjam Shows AI can amplify noise faster than real users care. Citation count ≠ traffic